Both executive functions in working memory (e.g. attention and response inhibition) and semantic functions are known to be compromised in schizophrenia. The former has been associated with general cognitive deficits, while the latter, more specifically with thought disorder and language use in schizophrenia. Currently, an ongoing debate in the literature pits the organization of the semantic system against executive functions as underlying the language disorganization seen in persons with schizophrenia. However, recent behavioral, physiological and imaging evidence suggests that language deficits, specifically thought disorder, are likely to be a result of interactions between executive processes and the semantic system. Accordingly, the long-term objectives of the proposal are to specify the interactions to the extent that they may be used for Identification of a biological and behavioral system useful for the early diagnosis of schizophrenia, as language abnormalities are often one of the first indicators of the disease. The semantic system interacts with executive processes at sensory, phonemic, word, phrase, sentence, and text levels. These interactions increase in complexity and thereby increase cognitive demands as the context moves from words to text levels. The specific aims of this proposal are to evaluate the interaction of executive and semantic processing during manipulations of contextual complexity, and potentially demonstrate both the relationship between behavioral performance during semantic tasks to other measures of semantic processing and the underlying brain system involved. We will specifically examine the effect of executive processes on semantic retrieval and examine the idea that in schizophrenia the failure of fronto-posterior connectivity, with a resulting compensatory increase in fronto- frontal connectivity, are part of language disorganization. This will be done using recently developed behavioral measures of semantic performance and semantic organization, in tandem with functional Magnetic Resonance Imaging and modeling to specify the biological system associated with our results. Healthy volunteers, persons with schizophrenia and their relatives will participate. Our behavioral and modeling results will increase the understanding of the mechanisms underlying semantic deficits in schizophrenia, which is of significant concern in the domain of public health, and has the potential to be an early marker of the disease.